AOpen Debuts The Funniest Motherboard Ever

Anonymous Coward X-11 writes "Has AOpen gone flipping nuts by putting vacuum tubes on its motherboards? AX4B-533Tube No, it's not replacing logic ICs with discrete components. The tubes are part of the on board audio. Not sure if they are serious about this. April 1 was two months ago." As an owner of a tube headphone amplifier I applaud AOpen's move to accomodate the high-end audio enthusiast, while simultaneous wondering about the ability of a switched psu to properly drive a tube amplification stage cleanly. There's no way this is for real, right? Right? Here's a link that seems to work pretty well. And this looks pretty, well, real. Update: /. reader Jedi1USA noted that HardOCP has more pics of the board.

Comments

who are they kidding?

[hifizen]

As an audiophile and electrical engineer, I can tell you that no serious music (tech) lover will take this seriously. It's common knowledge amongst audiophiles that high quality audio can only be reproduced outside the electrically noisy environment inside a PC case. But more than that, audiophiles are a particular, fussy lot, and they all have their own preferences for this tube type or that.

Who are you kidding, AOpen? Leave the high-end audio to the specialists, and leave it off your mobo!

Websurfing done right! StumbleUpon

Re:who are they kidding?

[User 956]

I can tell you that no serious music (tech) lover will take this seriously.

Most musicians are not engineers, and vice/versa. (I happen to be both, so I'll be glad to translate). Musicians are interested in sound production, and everyone else in sound REproduction. These are two totally unrelated things. A guitar amp is SUPPOSED to add distortion and tonal coloration to the signal. In fact tests for frequency response and distortion as applied to Hi-Fidelity equipment mean nothing when done on a guitar amplifier. We WANT to modify the original sound. That's exactly what the designers at Martin were doing from day 1. An acoustic guitar has a resonating chamber, a sound hole, and a front and back surface that act as diaphragms. If we could plot the frequency response of a single string played back on a good acoustic guitar, we would see all kinds of nastiness. Violins... the same thing. Saxophones... ditto. Pianos... of course. The sound quality of an instrument, what we call timbre (pronounced tam'ber) is a highly desired characteristic. It's what separates the Steinways from the Wulitzer uprights and the Martins from the Hondos, you get the idea.

In an electric instrument, there are no resonating chambers to add flavor, so we add it with distortion instead. Distortion, the crux of the biscuit..... sometimes we want it, other times we don't. That's what always bugs engineers and audiophiles alike. When you're trying to REproduce an already distorted sound, you don't want to add any additional distortion from the amplifiers. And you certainly don't want to add transistor distortion products. Sometimes tube distortion on a clean signal is OK. What, did I just contradict myself? No way, Jose'.

Here's the point of this whole exercise:

TUBE AMPLIFIERS SOUND DIFFERENT FROM TRANSISTOR AMPLIFIERS PRIMARILY WHEN THEY DISTORT!

You can see it on a o'scope, hear it with your ears, and prove it on paper, so I know it's true. And here's why: Transistors and tubes use different principles of physics for operation. Valve amps (the original name for tube amplifiers) are basically voltage driven, so when they distort, even-order harmonics are produced (2nd, 4th, 6th, etc...) while transistor amps are current driven and produce odd-order harmonics (3rd, 5th, 7th, etc....) When you look at them both on an oscilloscope, the transistor amp flattens out the waveform, while the tube amp sort of makes a triangle wave. If we look at the inherent resonances of acoustic instruments we find that things like violins make lots of even order harmonics, while things like clarinets make lots of odd order harmonics.

Now, do we want our guitar to sound more like a violin, or a clarinet? Ah yes, the violin is much more "sweet" sounding. That's because the human brain likes and will tolerate much more even-order distortion than it will odd-order. One of the great distortion boxes in recent history is the Aphex Aural Exciter. The guys who designed this thing were marketing geniuses. The original units were sealed boxes with tamper proof hardware. They were leased to studios for final mix down. Once a few big artists like Linda Rondstat and Fleetwood Mac used and liked them, their place in audio gadgetry was assured. Of course they just use a clever method of picking out the fundamental note of say a human voice, making a little distortion, and adding it back to the original signal. This adds "warmth" or "depth" to the sound that can be, well, exciting. Our theory about odd vs. even distortion holds up because while the first units made lots of odd harmonics, several years later I remember an engineering release that hyped a new distortion circuit with increased the 2nd order distortions, while reducing the 3rd order products. Sound's very interesting doesn't it? They made a way to produce "tube" sound in a device that's basically used on vocals and strings. Again, the brain likes even-order distortion while finding odd-order somewhat irritating.

There's some tricks to be used for making distortion. You can get transistors that are voltage driven called MOSFETS. In my opinion, they can be made to sound like tubes if properly designed. Also, once a signal is digitized, it can be made to sound like anything in theory, if the proper algorithm is applied. The problem is, they've been working on the sound of a Strat going through a Marshall for a long time. It's hard to beat the sound of a well designed tube amp, but they're beginning to learn how. Some of the new Multi-EFX boxes like the Quadraverb do an amazing job at processing a guitar. But I've found that they still need to go through a speaker to smooth out the sound. (There's additional distortion, coloration and bandwidth limiting produced by all loudspeakers)

Now for all you guys and gals who like their transistor amps and fuzz boxes, there's nothing wrong with what you do. Sometimes odd-harmonics are needed to cut through the mix. I sometimes purposefully loop a guitar track back into an overdriven channel on my board just to "dirty" it up a little. It can make a bland lead really stand out at times. But it's just a matter of taste. I really like the even-order distorted, bandwidth limited, compressed sound of a traditional tube amp for guitar.

But REproduction is a different story. I've got preamps that are rated in parts-per-million distortion. I tweak all my tape decks for the flattest response possible, and work hard to get the best signal-to-noise ratio possible. That's because I like the final sound to be as faithful to the original as possible.

audiophiles rejoice!

[Moosifer]

As a once snobbish, now reformed ex-audiophile, I cannot resist but slip back to the affected, bombastic days of my youth and exclaim "It's about freakin' time! Now when are we going to replace these markedly inferior CD-ROMs and DVD-ROMs with the gloriously mellifluous LP-ROM?"

New disk drive also in development for this board.

[DaedalusLogic]

It uses physical forms actually cut into a vinyl disk to reproduce sound. Rather than a traditional 5 1/4 disk bay it will be a USB 2.0 periperal and look something like these devices. Analog audiophiles rejoice! I play guitar and I do have to say I enjoy the sound of a classic tube amp. I wonder how odd and expensive will it be to say... Uh yeah I need a vacuum tube replacement for my computer.

Still, could be interesting for your HTPC

Re:Don't see how it's possible....

[jhines]

No, that is for a CRT, standard audio tubes work on a few hundred volts.

And it is the plate of a tube that gets the B+ high voltage, the grid regulates the current flow, and uses a modest voltage.

A tube amplfier has 3 power supples:
A- to power the filaments (6 or 12v AC)
B- Hi voltage DC, in the 300 volt range
C- Low voltage DC, to power the grid circuits

Why your dad says that...

[fmaxwell]

Tubes tend to produce even-order harmonics when they distort. Transistors (except MOSFETS -- others?) produce odd-order harmonics. Of the two, odd-order harmonics are much more annoying from a psycho-acoustic standpoint and lead to what many describe as a "harsh" sound. Tubes also have the advantage of not clipping hard (producing a DC output) because they have to run through transformers to drive speakers and, as we all know, transformers don't pass DC.

That's a gross oversimplification that leaves out much that I know, some that I think I know but don't really, and stuff other people actually know that I don't know at all. But that's the gist of it.

Is it of any use on a motherboard? Sure. It's great gimmick to sell to idiots. So how do they get stereo out of a single tube? It looks too small to be the two-tubes-in-one variety.

Well...

[cirby]

Compared to a high-end Pentium IV, a vacuum tube probably counts as a heat sink...

They're all harmonics!

[Mr Z]

Absolutely, the thing is that tubes distort harmonically while solid state amps tend to distort rather dissonantly and horribly.

Huh? They're all harmonics. Tube and certain kinds of FET (field effect transistor) based amps have a "soft limiting", so when they get close to clipping, they tend to generate even harmonics. Three of the first four even harmonics are exactly 1, 2, and 4 octaves away (2nd, 4th, and 8th harmonics), and so this form of distortion tends to be more melodic and pleasant. The 6th, 10th and so on aren't so melodic, but since the amplitude of the harmonics drops as you go to higher harmonics, you're ok.

BJT (bipolar junction transistor) based amps (and other types of solid state amps) tend to clip rather hard. No soft-limiting, they stop right at the rails. This clipping action creates a boatload of odd harmonics. These harmonics are fairly dissonant, giving the harsh sound most people complain about.

But they're all harmonics.

--Joe

Re:They're all harmonics!

[mesocyclone]

Errr.... harmonics at clipping are all fun and that, but only if you have one sine wave in your amplifier. With real music, you have lots of signal in there. Then, any kind of nonlinearity, whether tube or solid state, will produce intermodulation which consists of sum and difference frequencies. Intermod is infinitely annoying and the real problem with distortion in amplifiers.

The math of this is simple - and applies to RF as well as AF. Take two of the signals in the systemm and approximate them for as sin waves. The nonlinearity can be modeled as a power series, so you have terms of the form:

f(a,b) = A*(a + b) + B*(a + b)^2 + C*(a + b)^3...

Substitute

a=sin(w1*t+phi)
b=sin(w2*t)

And do the trig and you can see that you end up with all sorts of neat frequencies such as

w1-w2, 2*w1-w2, etc.

Now, instead of f(a,b), imagine f(a,b,c,d,e,f,g...)_ and you can see the mess that intermodulation makes. It basically mixes (in the frequency domain) all of the signals AND all of their harmonics in all possible combinations!

Tube amplifiers *do* sound different because their distortion curves are different than solid state amps. Why audio "purists" prefer one distortion curve to another is what I don't understand. What I want is minimal distortion overall!

But then, audiophiles also buy gigantic cables because they imagine that their speakers will sound better attached to them... etc.

Technical note: The coefficients on the various terms of the power series tend to go down with the order of the term. And, some configurations approximately cancel out all odd or even terms.

Re:My dad says...

[norton_I]

Well, there is a difference between a guitar amp and a stereo amp. For a guitar, the amp is really part of the instrument. Tube amps definately do mess with the sound, but the player desires that to get the sound they want.

For stereo use, the goal of the amp is (or should be) to reproduce with the highest fidelity possible the sound as recorded. There, while tube amps tend to be better than "cheap" solid state amps, a high end SS amp is at least as good as a high end tube amp.

I do know people, however, who prefer tubes on their stereo for whatever reason (mostly snobbery, IMHO). A lot of audiophile companies make ones with tube pre-amps and a SS final stage and/or all tubes. I haven't found these systems to sound particularly better than some high end SS amps.

Re:Only one tube?

[blair1q]

It's lower down on the page.

The tube is a dual triode, basically two tubes in the same package. It's got one or two cathodes, likely a common heater filament, two grids and two plates. The input signals go to the grids and the output signals are taken from the plates. Common cathodes would be fine, they'd both be connected to the same supply anyway. A common filament is an advantage, because it ensures the cathode(s) get the same heat, since thermodynamics plays a significant role in tube operation.

Answer your question?

--Blair
"Reading is fundamental."

Pictures

[CaptainSuperBoy]

at HardOCP

Re:Pictures

[n6mod]

Ahh, excellent link. From the photos, we can see that the tube is a Sovtek 6922. Google tells us that thetubestore.com calls the 6922 "Sovtek's premium version of the 6DJ8." Which everyone should recognize as a dual triode. Data here tells us that it has 6.3V filaments (perfect for coexistance with any 25120's that may be required by future DRM schemes), wants 90V B+ (plate voltage) and has a max Plate Dissipation of 1.8W.

This is a preamp tube, which is appropriate, actually. Get those nice GloFET harmonics and then feed it to some decent SandAmp for actual power.

Interesting concept. Twisted, and I don't think I want my preamp tube in the same faraday cage as my P4, but it is interesting.

-Z

Re:Oh that's what I need...

[Phork]

umm, there are water cooled tubes, uasualy they occur in RF amps, some older FmM broadcast stations used water cooled tubes in their finals. Water cooling a tube will let it work at a higher current than it is rated for. Basically the water cooling is to keep it from getting any hotter than it is supposed to get. The water cooling doesnt take it down to 50 deg F, it keeps it at the ideal operating temperature.